In the proposed research, the PI will: 1) come to thoroughly understand existing models of drug-behavior interactions in behavioral pharmacology. This will be done by (a) deriving them from more basic principles, where that is possible; (b) cataloging their applications, boundaries, strengths and weaknesses; (c) gauging the psychometricproperties of their parameters; their dimensions, sensitivities and intercorrelations. (2) Interrelate the existing models with Killeen's (1994) Mathematical Principles of Reinforcement (MPR) theory. (3) Develop appropriate tests of these interpretations and execute them. A series of experiments will determine if models developed from Killeen's theory can differentiate drugs based on these factors. Behavioral pharmacology is replete with hopeful attempts to quantify drug-behavior interactions. Examples 'include Hermstem's matching law/hyperbola, behavioral economics, behavioral momentum, delay discounting and signal detection theory. These and other techniques, procedures, theories and models are useful in describing phenomena at a local level. When they are viewed globally, however, it is unclear whether and how they are related to one another, or to the data just outside their domain. There is a need for a general theory of behavior that both incorporates the established correspondences between empirical observations and current theories, and provides a framework for the development of new models, of both local and global utility. Whereas existing models potentially could embody these characteristics, success is not guaranteed; meanwhile there exists one theory that has yet to be exploited to these ends. Killeen (1994) developed a quantitative model of behavior that was based upon three mechanisms; response constraint (ability), arousal (motivation) and coupling (short-term memory). Responses are energized by incentives, directed by coupling (more recent responses receive more of the weight of reinforcement than earlier responses), and bounded by temporal and physical constraints. These three mechanisms can be used to generate predictions about behavior under the influence of drugs. I propose to 1) derive several existing models from the theory; 2) generate new models based upon the theory; 3) test these interpretations and extrapolations with appropriate experiments. This process should help me perfect my education as a behavioral pharmacologist, deploy those skills to further research on problems of substance abuse and participate in the development of a comprehensive framework for the scientific modeling of the control of behavior by drugs.